Glutathione S-transferases (GST) are important in the detoxification of xenobiotics, catalyzing the nucleophilic attack by the thiol group of glutathione on the xenobiotic substrate. Since they catalyze the inactivation of several known carcinogens. Since they catalyze the inactivation of several known carcinogens, these enzymes can provide a defense against carcinogenesis. On the other hand, the elevation of GST levels in solid tumors appears to be a major factor in the development of resistance to treatment with cytotoxic agents. The GSTs are grouped into at least six different gene families based on sequence similarity and substrate specificity; and these isozymes differ in their ability to confer resistance to particular anti-cancer drugs. The amino acid sequences are known for the major dimeric mammalian GSTs and three dimensional structures have been determined for crystals of the pi-class, of the 303 isozyme of the mu-class and of the 1-1 isozyme of the alpha-class. However, important questions remain: (1) which amino acids contribute to the specificity of binding of xenobiotic substrates in the various GST isozymes; (2) does a given enzyme have more than one type of xenobiotic substrate site plus other non-substrate (possibly regulatory) sites; (3) which enzymic amino acids are the most important determinants of subunit interaction, as well as what role in the function of the enzymes i played by their dimeric structure. We will examine rat isozyme 1-1 as representative of the alpha-class of GSTs, rat enzyme 3-3 as representative of the mu-class, and pig lung enzyme as an example of the pi-class. These isozymes differ in substrate specificity and comparison of their sequences reveals 79-89% identical plus similar residues within a class, but only about 40% between classes. Our studies of the active sites of these enzymes while in solution will be complementary to and will be compared by computer modeling to structures of the protein crystals using the X-ray coordinates. We plan to use affinity labeling to effect specific modification and identification of amino acids in the xenobiotic substrate and non-substrate sites. We will use site-directed mutagenesis to replace amino acids proposed as participating in subunit interaction, as well as to evaluate the function of amino acids identified by affinity labeling. Mutant enzymes will be expressed and purified, and their monomer-dimer distribution, catalytic and binding characteristics will be examined. This study aims to provide the knowledge base for rational design of inhibitors specific for particular xenobiotic substrate sites for GST for use in novel combination chemotherapy to enhance the efficacy for alkylating cancer drugs.